recruiting and developing HPC staff
Yet, the number of these appointments that worked out well outnumbered the failures by fi ve-to-one. The cost of transferring an employee
from a non-HPC technical job is not especially burdensome. But it can take two years or more to become productive in the HPC environment. The most important task is preparing
people to exploit highly-parallel HPC systems. These systems pose enormous challenges for existing algorithms and applications. Although the HPC community has been delaying the inevitable, sooner or later (respondents disagree on when) the sustained-to-peak performance ratio on these systems will become so low that the HPC community will need to take action and rewrite algorithms and application codes. Surrounding HPC systems is a
‘complex’ of interrelated issues that need to be tackled together. These issues include not just extreme parallelism, but also petascale/exascale computing, HPC system heterogeneity, HPC system architectural balance, HPC system reliability, and HPC system and datacentre power and cooling. Expanding the markets for HPC could
make it a more attractive career path. HPC has been one of the fastest-growing IT markets and is a €13 billion worldwide market today. Further expansion could spark investment and make HPC an even more vibrant, attractive market for young people to enter. The single biggest recommendation
The candidate shortage affects the ability
of HPC centres to staff up fully, but it has not yet seriously affected the quality of their support staff. Only seven per cent of the centres characterised their support staff size as ‘very adequate’. The vast majority (93 per cent) of the centres rated support staff quality as either ‘very adequate’ or ‘somewhat adequate’. The best sources of job candidates are
universities and other HPC datacentres. The most fruitful source of qualifi ed candidates for HPC positions are ‘university graduates in mathematics, engineering, or the physical sciences’. The most productive non- academic source of qualifi ed HPC job candidates were ‘employees of other HPC datacentres’. Qualifi ed job candidates
from non-HPC centres are hard to fi nd, but often work out well. The least productive hunting ground for qualifi ed HPC positions was ‘employees at commercial or general IT datacentres’.
www.scientific-computing.com
was for universities to expand coursework in computational science – as opposed to mere computer science – and to integrate computational science methods into the requirements for science degrees, certainly at the graduate level and preferably also at the undergraduate level. Because HPC- based modelling and simulation has fi rmly established itself as the third branch of scientifi c inquiry, alongside theory and experimentation, respondents argue that HPC should therefore be as much a part of scientifi c education curricula as the two more established methods. HPC has contributed enormously and increasingly to scientifi c progress, industrial competitiveness, national security, and the quality of human life. But HPC advances are unattainable without an adequate number of properly trained personnel, including computational scientists, programmers, system administrators, technologists and all the others who help make up the HPC ecosystem.
and Life Sciences, and interim director of the Argonne
T
he talent pool working at the high end of HPC
computing is fairly limited, and we actively compete with
other laboratories and institutions to recruit and retain staff at the Argonne Leadership Computing Facility. We provide ongoing specialised
training to both new and existing system administrative staff due to the unique resources found in the facility and our commitment to maintaining the highest standards of safety. Our greatest challenge is to identify
individuals who have both domain expertise and an understanding of what it means to do computational science at scale. These individuals are invaluable to our mission, and I’m pleased to say that Argonne has been able to attract some of the best in the world.
OUR GREATEST
CHALLENGE IS TO IDENTIFY INDIVIDUALS WHO HAVE BOTH DOMAIN EXPERTISE AND AN UNDERSTANDING OF WHAT IT MEANS TO DO COMPUTATIONAL SCIENCE AT SCALE
The opportunity to work at a leadership
computing facility, a great work atmosphere and Argonne’s proximity to Chicago are all contributing factors. We actively cultivate a strong sense of community and work hard every day to maintain our worldwide reputation as a premier facility dedicated to open science. Gaining access to machines and
experiments at the cutting-edge of computation is what attracts a wide range of HPC professionals to the candidate pool. Supporting scientists in their pursuit of scientifi c breakthroughs is what helps retain them.
AUGUST/SEPTEMBER 2011 37
associate laboratory director for Computing, Environment
Michael Papka, deputy Leadership Computing Facility
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48